S3 Transfer
Browse, upload, and download files from S3-compatible object storage using the dual-pane file browser.
S3 Transfer mode connects to any S3-compatible object storage service — Amazon S3, Wasabi, MinIO, Cloudflare R2, DigitalOcean Spaces, and others. The right pane becomes an S3 object browser with familiar file-system-like navigation.
Setting up an S3 connection
- In Connections, click New Connection.
- Switch the Connection Type toggle from SSH to S3.
- Fill in the S3 fields:
| Field | Required | Description |
|---|---|---|
| Name | Yes | Friendly label for the connection card |
| API URL (Endpoint) | Yes | S3-compatible endpoint, e.g. https://s3.amazonaws.com or https://s3.ap-southeast-1.wasabisys.com |
| Region | No | AWS or custom region (e.g. us-east-1, ap-southeast-1) |
| Bucket | No | Default bucket — leave blank to browse all buckets first |
| Access Key | Yes | S3 access key ID |
| Secret Key | No | S3 secret access key (stored encrypted in the vault if saved) |
| Group | No | Organise alongside your SSH connections |
- Click Test Connection to verify your credentials and endpoint before saving.
- Click Save Changes (or Connect & Save from the Quick Connect dialog).
Leaving the Bucket field empty allows you to browse all buckets on the account first, then pick one to descend into. This is ideal when you manage multiple buckets under the same credentials.
Starting an S3 session
- Double-click an S3 connection card in Connections, or
- Right-click the card → Browse S3.
The right pane connects to your S3 endpoint and lists the contents. If you configured a default bucket, it opens directly. Otherwise, it shows a list of all buckets — click any bucket to browse inside.
The bucket:// path scheme
When connecting without a default bucket, the address bar uses the special bucket:// prefix:
bucket://— list of all bucketsbucket://my-bucket— root of the bucketmy-bucketbucket://my-bucket/uploads/— folderuploadsinsidemy-bucket
Navigating up from bucket://my-bucket/... returns to the bucket list, not a broken "parent" of /.
Transferring files
Upload (local → S3)
- Navigate to the target folder in the S3 Remote (right) pane.
- Select one or more files in the Local (left) pane.
- Drag them to the S3 pane or right-click → Upload to S3.
Download (S3 → local)
- Navigate to the target folder in the Local pane.
- Select files in the S3 Remote pane.
- Drag them to the local pane or right-click → Download to Local.
Large file handling
Files larger than 1 GB trigger a confirmation dialog before queuing. The dialog warns about:
- S3's 5 GB single-transfer limit — files exceeding this will fail.
- No resume on cancel or network drop — if a transfer is interrupted, all progress is lost and it must restart from the beginning.
For very large files (over 5 GB), consider using a dedicated S3 client with multipart upload support.
Background queue & progress
All transfers are queued and processed in the background, exactly like SFTP:
- Navigate away or switch tabs while transfers run.
- Monitor progress in the Activity Panel (`Ctrl+``).
- Cancel an in-progress transfer from the queue (cancellation is checked between chunks, so large files respond quickly).
- The max parallel transfer jobs setting in Settings → File Transfer applies to S3 transfers too.
Conflict handling
When a file already exists at the destination, Ternux shows the same conflict dialog as SFTP:
| Option | Behaviour |
|---|---|
| Overwrite | Replace the existing file |
| Skip | Leave the existing file untouched |
| Rename | Write the new file with an auto-incremented suffix (e.g. file-1.txt) |
| Cancel All | Stop the batch and keep all existing files |
File operations
Right-clicking a file or folder in either pane exposes:
| Action | Description |
|---|---|
| Open | Navigate into the folder (or select a file) |
| Download / Upload | Transfer the file |
| Edit | Open the file in the text editor dialog |
| Rename | Rename the file in-place (uses S3 copy + delete) |
| Delete | Permanently delete the object |
| New File | Create an empty object at the current path |
| New Folder | Create a zero-byte folder marker |
| Copy Full Path | Copy the display path to the clipboard |
Folder rename is not supported on S3. S3 has no real directories — renaming a folder would require copying and deleting every single object inside it, which is slow and can leave data in an inconsistent state if interrupted. Create a new folder and move files there instead.
Delete in the S3 pane permanently removes objects. For folders, all objects under the prefix are deleted recursively. There is no trash or undo.
Pagination & infinite scroll
S3 buckets with thousands of files are loaded incrementally. As you scroll to the bottom of the S3 pane, the next page of objects loads automatically — no "Load more" button required.
Compatible providers
S3 Transfer mode uses S3's standard API and has been tested with:
- Amazon S3
- Wasabi (path-style addressing used by default)
- MinIO
- Cloudflare R2
- Any S3-compatible service
If your provider requires a specific endpoint, enter it in the API URL field when creating the connection. The app automatically uses path-style addressing (https://endpoint/bucket/…) for maximum compatibility.